Brian Hulette

Hi @imadmoussa1, according to the Python docs [`signal.pause` is unix only](https://docs.python.org/3/library/signal.html#signal.pause). Are you running on a non-unix environment? (note we _should_ support, so if that is the issue this is...

It looks like that might be an issue in the Python expansion service, which we should track down (the full stacktrace for the signal.pause error would help here). But that...

> @robertwb I think you had concerns about this change. > > Do you think this should be reverted ? I'm curious what the concerns were. My understanding was that...

Run Python 3.8 PostCommit

Run Python PreCommit

> @TheNeuralBit ptal and lmk if the io_it_test looks viable Looks great! I can merge when we get CI green

I'm not sure that's the root cause of the timeout, I've seen errors like that before, they typically are the root cause of an actual test failure. In this case...

No I mean for all Python PostCommits (your link is correct), since that's the timeout we're hitting. Per @kennknowles - timeouts are for catching runaway jobs, not for punishing us...

Yeah I agree that we should do something to address the long-running PostCommits after this is resolved. We've had some discussions about splitting up the PostCommit like Java is doing.

I think the "No producer" error is a bug with PCollection pruning in interactive beam: #21430 CC: @KevinGG